UB/Bet21 Seating Assignments: feeding the conspiracy theorists

Huh?

These stats look very close to me. The only one out of expectancy, in your sample, is the dealer busting which tends to say you need to play this game very aggressively according to the outcomes.

 

Ha ha ha. Isn't it funny that people who DON'T do something are ALWAYS first to give advice about doing something?

Yes, these raw number do look very close to the expectancy at first glance. But there’s more to the story.

These dealer hands are over 40 different games.

In some of these games the results are so skewed as to be ridiculous. For example:

21 hands:
Ace up 3 times (14.3%)
Dealer BJ 2 times (9.5%)
Dealer draws 19-21 (w/o BJ) 13 times (61.9%)
Dealer busting 3 times (14.3%)

That was my most extreme example. In fact there are certain times when I would have a 13 or 14 with the dealer showing a 2 or 3. I “just knew” if I stood (basic strategy) that I would loose. So I would hit those hands and sure enough I take the card that would have given the dealer a 20 or 21.

Now we’ve all seen that at one time or another. But when it happens once every other game you’ve got to get suspicious. In fact it has begun to “psyche” me out and I don’t play in the same manner that I used to play.

The other thing that “raw numbers” don’t tell is when these events occur. From a cursory examination of the numbers the amount of dealer busts in the first 8 hands is much lower than in the second 8 hand sequences.

Therefore, using YOUR recommendation about overly aggressive play in the early stages with the way cards are currently being dealt you are at an extreme disadvantage.

With the pages and pages of data I could spend days or weeks sifting thru the numbers. The saddest part is that the first two (2) cards that I were dealt show a definitely high proportion of “stiffs” (12-17) hands than one would expect.

Now was this just me or all players?

Can you see where this is all leading? Conspiracy theories, paranoia, etc. TBJ is supposed to be challenging – yes. But it’s also supposed to be FUN. Right now Bet21.com ISN’T fun. I’m spending so much time collecting data that it feels like work and for what? To prove/disprove that the software has/doesn’t have flaws?

Like the idiots at Bet21.com even care! That’s what pisses me off more than anything else. They paid people to get the software right. They had beta testers and they still managed to screw it up. Why should I spend my time and money so that these same people who lied to me about the bonus money and who don’t communicate can enhance their product?

Nah, it’s time to take my marbles and go home. It was fun while it lasted but the party’s over. For those of you remaining – good luck

 

I agree

bet21 has problems with the dealing mechanism and has for a long time IMO, just look back under online casino for a thread I started "Anomaly?"

For the online casinos I have played goldenpalace dot com had the most natural cards, followed by globalplayer altho GP dealt low counts n average. BJ21 has somewhat natural dealings but tend slightly to the high side- which makes it a place that tends to help the aggressive player. Lets face it, random computer generations are wide open to unnatural trends. You just gotta take what is offered and adjust your play to fit.

 

Yes, but no.

Yes, the computer generations are NOT random and they are very unnatural.

No – you don’t have to take what is offered.

If a site has a bias for one way or another that is constant and consistent it can be dealt with. However, when the cards are constantly changing it makes for a very burdensome game.

For example anyone who observed me play in the past would definitely NOT categorize me as a conservative player. Nevertheless I found myself betting minimum hand after hand while I gathered data from Bet21.com This often benefited me on numerous occasions.

Last night while playing at the final table on BJ21.com on the final hand I bet minimum when I was BR2. I was last to bet. It was a TERRIBLE bet. But I have been so conditioned by Bet21.com to expect the dealer to draw to a 19-21 that I was convinced the same thing would have happened.

In reality dealer busts and I go from BR2 to BRL-1 (BRL had lost all their chips earlier).

Now 3 months ago I would have won that game because I would have been doing progressive bets and on the last hand I would have wanted to take the high. Now, I’m “gun shy” and for that one reason alone I’ve go to stop playing Bet21.com.

There are other reasons NOT to play at Bet21.com also.....

 

thx

Yes, I realized you all are I think talking about bet21 - but I do not play there - only UB - still end up in seat 5 a lot. But that is first round so it doesn't matter as much anyway. Thanks though (from a newbie) for something to ponder

 

eliburk

I seem to be responding to your posts

Ultimate Bet and Bet21 are the exact same games - Bet21 is just a skin of UB - so everything posted about either applies to both as far as the game itself and customer service issues go

 

Can I play?

FWIW I don't believe that there is anywhere near enough data here to conclude that Bet21 have skewed/rigged dealing software. But I can see why a random number generator may not be able to mimic the real-world card shuffling mechanism, be it CSM, card shuffler or human. Could it not be argued that in fact RNGs are more random that the real world applications and that this "pure" randomness is anomalous with the non-random live experiences that a lot of you have?

Can I ask a fundamental questions here?

1) Can someone define random for me?
2) How do you prove or disprove randomness?

Cheers

Reachy

 

Wiki Wiki

Reachy,

There's quite a lot of useful stuff on Wikipedia (with the usual caveat that it might all be a load of rubbish, I suppose ). If you start out by searching for 'random' or 'RNG' you can follow links to all manner of related items.

 

Stimulate

Colin, I have visited The Great Oracle Wikipedia but my questions weren't really posed to get answers, if that makes sense . I wondered whether people actually know what randomness is and whether they can truely evaluate whether an event or series of events is random or not. Just trying to get people thinking/talking and using fundamentals rather than viewing the problem through the red mist that is "The Bet21 Issue".

Cheers

Reachy

 

Deviation not right

It could very well be the algorithm is correct to yield the proper outcomes but the frequency of the changes is too slow. This would offer a person doing a study of a few hours a day for a few days to find anomalies. In real BJ you have different shufflers, aka dealers or machines, which make the changes of card order rapid. Maybe the RNG at UB is slow, offering obvious patterns that are actually just slow generations of the natural event. And so on:joker: But, I dont thinkyou can ignore the reputation of UB poker, which is for abnormal patterns as well.

 

:laugh: :laugh:

 

randomness

as I posted quite a while ago - from the description of the RNG UB uses for their poker games, and it is a very reasonable assumption this is also used for their bj and tbj games - they use a dual zener diode hardware RNG - this setup works by sampling the electrical noise generated by the diodes - mixes streams of random occurances, and uses a standard set of software algorythms to test those streams for randomness - this is not a software pseudo-random number generator - the randomness tests verify that there is no external factor that would bias the results and that the diodes haven't degraded -

these RNGs are cheap and effective - they cost less than $1,000 and are used in research labs -

the 'anomalies' you keep seeing are the EXPECTED results as the probabilities work themselves out over a series of short sequences and are not unusual - in fact - not seeing such 'anomalies' would be unusual and supicious -

it is well known that in any random number stream - you will find sequences which appear to be non-random - by random chance - these (apparently non-random) sequences themselves should show a random distribution - and all the tracking I have done on UB/Bet21 suggests this seems to be the case -

I am satisfied that the players' cards are suitably randomized, enough so I don't see the need for further tracking to convince me of this - at least until I see something to suggest otherwise, and I haven't yet -

'RANDOM" does not mean "matches the projected results", it means RANDOM

 

"I am satisfied that the players' cards are suitably randomized, enough so I don't see the need for further tracking to convince me of this - at least until I see something to suggest otherwise, and I haven't yet -"


I equally convinced and unsatisified that the cards are NOT randomized.

If I havent seen enough to prove that further tracking is necessary..how can you say the opposite? You offer no assurance other than ...Oh...it probably not happening...and you have no facts or stats to back an all is well attitude...

You are making an at minimum equally unsubstaned , and untested statement.

How can you be so sure? Dont tell me what you have seen .... tell me what have you proven?

 

Rando

I posted data on more than 4,000 hands tracked on the Boycot Bet21 thread - showing that the probabilities for the players hands were well within the expected range - as for my opinion - I do have a degree in math with graduate work - specializing in research methodology - which doesn't necessarily make me right - but it does suggest I know what I am talking about - and not just voicing a personal opinion -

the bottom line here is that everything that has been described as 'abnormal' results are very much normal results - which anyone with expertise in research methods and statistics would know - when you look at very short sequences, and the large numbers of tables played - you EXPECT to see this - and I would be very suspicious if I did not see these types of things - as that WOULD be abnormal - if they did not occur -

I have had my own paranoia about Bet21/UB - which is why I have done some extensive tracking of results - and why I searched out the info about their random number generator - but all the tracking I have done hasn't come up with anything out of line -

everyone keeps expecting every table to produce the 'expected probabilities' we know from the published probs based on infinite sequence calculations and very long simulations - it doesn't work that way - finite sequences work differently - short sequences can, and will, produce extreme results - that's just the way it works - it is the accumulation of short sequence results - that converges to the expected probabilities over the very long term - and that very long term means tens of millions of hands - not a few thousand -

to date no one has produced any data that suggests that the cards/hands dealt by UB/Bet21 are abnormal - and that is my professional opinion as a person with expertise in math/statistics, who has worked as a consultant/advisor/instructor in research methods and statistical analysis - and done statistical studies and modeling - as well as my personal opinion as a player -

 

damn; i struggle, at times to count to 21.
great information, and well stated!!!!

 

Please... your resume doesnt prove anything...nor did I ever question you knowledge on the subject...

But in fact I am using your exact same argument only to the opposite position...

You havent run anything near what could be considered a conclusive test to make an assumption that everything is ok. Yet, you expect me to take your word and further your list your resume as some proof that the numbers are ok...when in fact there is no way you can say (with such a small sample) that nothing is suspect...

I can spout off ideas as well but without tests they are only words w/o meaning.

You may understand that 2 plus 2 = 4 but you have offered nothing to prove that. You have nothing to prove everything is ok except for an assumption that it is so....that plus a too small sample (by your own word) to be meaniful...

You cant have it both ways..... Its either unknown and untested and therefore suspect... or you have run the proper test and are assured...

You are guessing....nothing more. And your education may make that an "educated" guess but the guess status remains....

Is that math or fuzzy logic?

I played 6 rounds tonight of MTT.....In the last two games I had Bj that was tied with the dealer BJ....I say this is rare occurance...but I also saw the dealer tie another players Bj two times in the same game...making it twice as rare...yes of course possible....but i know I will see that tomorrow night as well...I will be looking for just that....but i know I will se it....Im fishing yes.....But I bet you or anyone else I will "catch that fish!


I say the dealer gets an abnormal number of dealer ACE up....I have been watching (fishing ) for that for 6 months....Im still seeing it far more than the the statistics allow as a norm....what is it like 3 - 4 % of the time a dealer should show a ACE up....???

When it happens 3 out of 8 hands for 65% of all games I play then I no longer feel comfortable that randomness is happening...

Im catching this same fish every day for months now....guess what? its possible this lake is full of this type of fish....and because in 3,000 casts you havent caught one yet....does that prove that this fish does not exist? No .

You can not give an all clear....so stop trying until you can show the test that proves it. And you certainly know and understand the rules for that far better than I do ....I studied accounting and a few stat courses but science was my thing...I could be exactly wrong....but Im gonna need proof at this point ...not just small sample based assumptions....I have already made those myself...;-)

 

Babbage

I always read RKuczek's posts with interest and due to his background in maths and statistics I tend to believe them to be accurate. My personal knowledge is limited academically to school maths but I was reasonably talented and have always slightly regreted not taking it to degree level and further. But that is 20 years ago now .

What I am saying is that I can still understand complicated mathematical/statistical proofs if they are explained to me. What I'd like RKuczek to do, if he feels inclined, is to explain, in mathematical rather than laymans terms, why he believes Bet21's RNG to be producing a truely random "deal". If I don't understand something in the "proof" then I will ask.

Just so you know where I stand, I haven't yet seen any convincing or statistically significant evidence that Bet21 have a non-random RNG posted on this site. That doesn't mean that it isn't dodgy, it just means I haven't seen anything to convince me of this fact. My gut reaction is that in fact people are seeing a "true" random process rather than the psuedo-random shuffling process they may be more used to in B&M casinos.

Anyway, what do you think RKuczek, will you shows us your evidence?

Cheers

Reachy

 

Perspective

There undoubtedly hasn't been enough data collected to refute a claim of non-randomness in the cards, but the real question is - has there been enough to warrant making the claim in the first place?

I've experienced the same - "I don't f*#!ing believe it! This game must be rigged." - response to wild downward swings in my fortunes at various casinos, both online and real-world. I'm sure I'm not alone in that.

In fact I personally haven't really noticed/imagined anything particularly out of line (aside from the seating assignments) at Bet21, whereas elsewhere I have. Perhaps I'm just not paying attention.

So if what has been seen so far falls within the bounds of what can be considered normal, then if there is a reason to seek assurance that all is well with the cards at Bet21 it stems more from the general mistrust of that company than from what has been observed at the tables.

Mistrust needn't mean that you think the game may be rigged, it may just be that you worry they don't take enough care to ensure everything is working properly.

This is the real cost of poor customer service!

 

London,

Excellent post. Take the final hand that TedinNaples posted last night. With a 90+% chance of advancing to have the dealer draw to 21 sucks. Does it happen in B&M casino's? ABSOLUTELY.

The problem here is the lack of customer support/communication.

With regard to abnormalities with cards and randomness issues.

In Las Vegas they have a gaming commission that checks the chips on all the slot machines to ensure a "random" number generator.

A few years ago one of the people who worked there, found a flaw in a chip associated with Keno. He exploited it and made millions - until he went overboard and spent some time in the pokey!

For example we know what the odds of a dealer busting are - 28.2%. Using a simple binomial calculator we find that the probability of a dealer only busting 1 time in the first 8 hands is 22%.

At what point do red flags go up? At what p level does it become significant?

I don't know - I really don't. But I know this much, in a B & M casino I see the cards and see a shuffle. I wouldn't play in a TBJ contest with a Spanish 21 shoe (missing the 10's) and if I did boy would I be surprised at the results.

Lets look at the TedinNaples case. Bad beat? for Sure. Dealing abnormality? Well if you put that ONE hand in with 999 others it dilutes the results so that in essence large numbers can "hide" a small anomoly. The question becomes is it a natural occurence or a synthetic or fabricated occurence?

Without a virtual gaming commision or indepenent certification we can't know for sure.

I guess it all comes down to trust.

 

Engineer

Ah, but do you? How can you be sure that the dealer isn't a mechanic?

On one of the poker forums I lurk at there are often threads along the lines of "abcpoker is rigged" or "montypythoncasino dealing is dodgy, I'm only going to play live games from now on" and one of the most common responses from the players experienced in both live and internet gaming is "it's the live games that are rigged". Obviously poker and bj are different games and cheating/mechanics is probably much easier in poker but you get my point....

Cheers

Reachy